Periodic $p$-adic Gibbs measures of $q$-states Potts model on Cayley tree: The chaos implies the vastness of $p$-adic Gibbs measures

TL;DR

This paper investigates the complex structure of $p$-adic Gibbs measures for the $q$-states Potts model on Cayley trees, revealing chaotic behavior and vastness of measures through symbolic dynamics and $p$-adic analysis.

Contribution

It demonstrates the chaotic nature of the Potts--Bethe mapping over $ extbf{Q}_p$ for certain primes and parameter regimes, establishing the existence of a large set of periodic $p$-adic Gibbs measures.

Findings

Existence of subsystems conjugate to full shifts on three symbols.

Identification of subshifts of finite type with at least four symbols.

Chaotic behavior of the Potts--Bethe mapping for $p eq 2,3$ and specific parameter conditions.

Abstract

We study the set of $p$-adic Gibbs measures of the $q$-states Potts model on the Cayley tree of order three. We prove the vastness of the periodic $p$-adic Gibbs measures for such model by showing the chaotic behavior of the correspondence Potts--Bethe mapping over $\mathbb{Q}\_p$ for $p\equiv 1 \ (\rm{mod} \ 3)$. In fact, for $0 < |\theta-1|\_p < |q|\_p^2 < 1$, there exists a subsystem that isometrically conjugate to the full shift on three symbols. Meanwhile, for $0 < |q|\_p^2 \leq |\theta-1|\_p < |q|\_p < 1$, there exists a subsystem that isometrically conjugate to a subshift of finite type on $r$ symbols where $r \geq 4$. However, these subshifts on $r$ symbols are all topologically conjugate to the full shift on three symbols. The $p$-adic Gibbs measures of the same model for the cases $p=2,3$ and the corresponding Potts--Bethe mapping are also discussed.Furthermore, for $0 < |\theta-1|\_p < |q|\_p < 1,$ we remark that the Potts--Bethe mapping is not chaotic when $p=2,\ p=3$ and $p\equiv 2 \ (\rm{mod} \ 3)$ and we could not conclude the vastness of the periodic $p$-adic Gibbs measures. In a forthcoming paper with the same title, we will treat the case $0 < |q|\_p \leq |\theta-1|\_p < 1$ for all $p$.